Skip to main content

Table 1 Continuous classification of allomaternal care behaviours: Averaged parameter estimates and their relative explanatory importance for female CV body mass (N = 87). Gestation length and neonatal mass were excluded to reduce multicollinearity between predictors. Numbers in bold indicate predictors whose confidence intervals of their effect exclude zero

From: Getting fat or getting help? How female mammals cope with energetic constraints on reproduction

Predictors

Relative importance of predictors

Model averaging estimatesa

95% CI

Intercept

  

0.126

(0.100, 0.153)

Provisioning

 

1.00

−0.040

(−0.043, −0.036)

Protecting

 

0.06

−0.001

(−0.002, 0.001)

Carrying

 

0.07

0.003

(−0.004, 0.010)

Communal nesting

 

0.06

0.001

(−0.002, 0.004)

Allonursing

 

0.06

0.005

(−0.010, 0.021)

Log mean body mass

 

0.44

−0.006

(−0.010, −0.002)

Provenance

captive

0.80

na

na

wild

0.025

(0.017, 0.032)

Substrate use

terrestrial

1.00

na

na

arboreal

−0.045

(−0.050, −0.041)

Number of months

 

0.53

−0.001

(−0.001, 0.001)

Inclusion of reproductive females

 

0.69

−0.019

(−0.030, −0.008)

Log litter size

 

0.56

0.027

(0.013, 0.041)

Log weaning age

 

na

0

0

  1. aaveraged model estimates based on 12 models with ΔAICc (AICc focal model – AICc best model) < 2 since the best AICc model is not strongly weighted (weight = 0.15) [104]. A full list of models is given in Additional file 4: Table S4. Reference levels of categorical variables have an estimate of 0; na – not applicable; 95% CI - 95% confidence interval